Monomeric and dimeric antibody-fragment fusion proteins

Abstract

The present invention describes a new class of antigen binding molecules which contain Fv-fragments of an antibody but do not use the constant antibody domains. They can also dimerize with other antibody fragment molecules or with non-antibody fragment molecules to form bi- or multifunctional antibody-fragment fusion proteins and so-called miniantibodies, respectively. The new fusion proteins can be used in the broad field of diagnostic and therapeutical medicine.

We claim: 
     
       1. A dimeric protein comprising two monomeric fusion proteins in a noncovalent interaction, wherein the orientation of the monomeric fusion proteins in the dimer is identical with respect to their N- and C-termini,   the noncovalent interaction between monomeric fusion proteins is formed between interactive peptides on each monomeric fusion protein, which peptides are identical with each other or nonidentical, and   each monomeric fusion protein is a monomeric antibody-fragment fusion protein comprising: (a) a single chain Fv (scFv) fragment of an antibody,   (b) a linker peptide, and   (c) a first interactive peptide comprising a first amphiphilic helical peptide containing up to 50 amino acids, said first amphiphilic helical peptide being capable of forming a dimer with a second amphiphilic helical peptide by noncovalent interaction, wherein the C-terminus of (a) is fused with the N-terminus of (b), and the C-terminus of (b) is fused with the N-terminus of (c) and,       with the proviso that if either monomeric fusion protein comprises a helical peptide which is a jun-zipper, the other monomeric fusion protein may not comprise a helical peptide which is a jun-zipper.   
     
     
       2. A dimeric protein of claim 1, wherein the interactive peptide in at least one of the monomeric fusion proteins of the dimeric protein comprises a helix, a turn and another helix. 
     
     
       3. A dimeric protein of claim 1, wherein the interactive peptide in at least one of the monomeric fusion proteins of the dimeric protein contains a leucine zipper molecule having several repeating amino acids, wherein every seventh amino acid is a leucine. 
     
     
       4. A dimeric protein of claim 1, wherein the interactive peptide in at least one of the monomeric fusion proteins of the dimeric protein bears charged residues. 
     
     
       5. A dimeric protein of claim 1, wherein the linker peptide in at least one of the monomeric fusion proteins of the dimeric protein is a hinge region sequence of an antibody or a fragment thereof. 
     
     
       6. A dimeric protein of claim 1, wherein the interactive peptides (c) on each monomeric fusion protein are identical. 
     
     
       7. A dimeric protein comprising a first monomeric fusion protein unit and a second monomeric fusion protein unit in a noncovalent interaction, wherein the noncovalent interaction between monomeric units is formed between interactive peptides on each monomeric unit, which peptides are identical with each other or nonidentical,   each monomeric unit is an antibody-fragment fusion protein as set forth in claim 1 and   the first monomeric unit and the second monomeric unit have different antigen-binding specificities.   
     
     
       8. A dimeric protein of claim 1, wherein another protein is fused at the C-terminus of one or both of the interactive peptides. 
     
     
       9. A process for preparation of a dimeric protein of claim 1, comprising cloning the DNA molecules having sequences coding for a monomeric fusion protein, or a part thereof, into at least one expression plasmid,   transforming a host cell with said expression plasmid or plasmids,   cultivating said transformed host cell in a nutrient medium, and either (A) expressing the dimeric protein in the cell, or   (B) separately, expressing the monomeric fusion proteins or a part thereof, and forming the noncovalent linkage between the two monomeric fusion proteins in the medium or in vitro, wherein when only a part of the fusion protein was cloned into one expression plasmid, the other part or parts of the fusion protein are assembled and fused in vitro with the cloned part to form the fusion protein, before or after forming the noncovalent linkage.       
     
     
       10. A process of claim 9, wherein the DNA sequence coding for the first monomeric fusion protein is cloned into a first expression plasmid, and the DNA sequence coding for the second monomeric fusion protein is cloned into a second expression plasmid. 
     
     
       11. A process of claim 9, wherein the noncovalent linkage between the monomeric fusion proteins forming the dimeric protein is formed in vitro. 
     
     
       12. A process of claim 9, wherein the host cell is E. coli. 
     
     
       13. A dimeric protein of claim 1, wherein the interactive peptides on each monomeric fusion protein in the dimer are nonidentical. 
     
     
       14. A dimeric protein of claim 13, wherein the interactive peptides on each monomeric fusion protein contain leucine zipper molecules, whereby the formation of heterodimers is favored. 
     
     
       15. A dimeric protein of claim 14, wherein the interactive peptide of one monomeric fusion protein contains a jun-zipper and the interactive peptide of the other monomeric fusion protein contains a fos-zipper. 
     
     
       16. A dimeric protein of claim 13, wherein the noncovalent interaction between the interactive peptides on each monomeric fusion protein is by intercalation. 
     
     
       17. A process of claim 9, further wherein, in step (A), the dimeric protein is secreted into the medium after expression. 
     
     
       18. A dimeric protein of claim 1, wherein the linker peptide is a hinge region. 
     
     
       19. A dimeric protein of claim 18, wherein the hinge region has an amino acid sequence selected from the group consisting of the hinge region sequences of SEQ ID NOS: 9-14.